4.3 Results and Analysis
4.3.1 Biographical data findings
46
• An insight into other sources of L2 interference other than the respondents L1; for example, television programmes learners watch; radio stations learners listen to; types of magazines and newspapers learners read.
• Evidence of L1 interference; examples on these sections were clarified in the discussion of the test instrument, which is given in the next section below.
With regard to the feasibility of the questionnaire and test as data collection instruments, the pilot study revealed that there was no obvious challenge as no setbacks were experienced during the pilot. Thus, learners did not show evidence of struggling to deal with the questions posed in the instrument. There were also no ambiguities associated with the questionnaire’s instructions and layout. The learners did not take a long time to finish responding, as the first learner completed the questionnaire within 40 – 45 minutes and the last one, just on 1 hour. The pilot effectively served as a guideline for the coding/classification system for data analysis. Furthermore, no parts were found to be too difficult, too easy, or impossible to follow. Hence, the test was considered the most effective measure for getting the most useful data in the least amount of time as it was envisaged.
After revising the instruments, the final research was conducted on Saturday, 28 November 2011. All the respondents were grade 12 learners at the time of data collection. Even though there were a total number of 65 learners in the grade 12 class in question, it was expected that only 58 would participate in this study. On the day of data collection, only 55 learners turned up to complete the questionnaire and sentence transformation test. The next sections present and discuss the results of the data collection for the study.
4.3 Results and Analysis
47
Table 4.1 below presented biographical findings of the learners’ gender, age, home language and the additional language spoken by learners other than their home language. The item findings are presented in calculated total numbers and percentages.
Table 4.1: Biographical Data
Item Total Percentage
Gender
Male 25 45.5
Female 30 54.5
Age:
16 -18 19 34.5
19 – 21 23 41.8
22 – 26 13 23.6
Home Language
Xitsonga 48 87.3
Tshivenda 01 1.8
Did not specify 06 10.9
Other Languages
Sepedi 01 7.3
Tshivenda 04 7.3
English 25 45.5
Xitsonga 01 1.8
Did not specify 24 43.6
Table 4.1 revealed the biographical data as follows:
4.3.1.1 Gender of the respondents
With regards to gender, the findings indicated that the total number of male respondents was 25, while the female respondents were 30. This represented a percentage total of 44.5% and 54.5% respectively. According to Cameron (1995), there are three models of language and gender, namely: the deficit model, the dominance model, and the cultural difference model.
48
In the deficit model, females are seen as disadvantaged speakers and communicators. Accordingly, the speech of men is considered as the accepted norm, while women’s speech is perceived as deficient. In the dominance model men dominate and control women and language. Men in this model control topics of conversation, interrupt more and talk more than women (West & Zimmerman, 1987). The third type, the cultural difference model refers to the different biological forms of language used by men and women. This difference is, according to Mabila (2001), due to their early socialisation and lead to different rates of language acquisition. Women in this model tend to attach more value to making connections, seeking involvement and concentrate on interdependencies between people, whereas, men value autonomy and detachment and seek independence, focusing on hierarchical relationships.
According to the researcher, the advantage in this study is that there were more females than males; therefore the researcher is of the opinion that the disadvantage pointed out by Cameron (1995) could not be applied to this study, since there were a slightly higher number of female participants.
The next section presents a discussion on the age of the respondents.
4.3.1.2 Age of the respondents
From Table 4.1 above, it can be deduced that out of the total of 55 learners, who participated in the study, 19 were between the ages 16-18, 23 fell within the 19- 21 age range and 13 fell within the 22-26 age range. The researcher noted with concern that most of the learners, who participated in the study were way above the expected grade 12 ages. Although this was a matter of concern, it was not, however, an issue for this study, except for the fact that these learners would have had a long period of exposure to the sources of language input.
49
The subject of age has often been considered a major factor, which determines success in learning a L2. Collier (1987) points out that when children are asked to learn a L2 for use at school before their first language has sufficiently matured to serve as a source of transferable skills, the learning task is very burdensome and requires more time than older children need. This means that the home language of the respondents affect learners’ studying of L2 both on a positive and negative level. Positively, learners use L1 structures, which are the same as that of L2, correctly. This view is supported by Sanderson (2005:45), who says,
“If the relevant unit or structure of both languages is the same, linguistic interference result in correct language production as aspects of the L2 and, the L1 is learnt more easily, because they do not have to be learnt from scratch”
(Sanderson, 2005:45). In addition, Dulay, Burt and Krashen (1982) as well as Ellis (1994) have noted that when writing or speaking the TL, L2 learners tend to rely on their L1 structures to produce a response.
From the age of the respondents in Table 4.1 it was observed that the respondents’ ages ranged between 17 and 26 years. At this age most learners have passed the critical period hypothesis, as such most of them have fully acquired their home language. The learners between these ages (17- 26) use L1 structures as a foundation in producing English sentences and, the L2 is learnt more easily, because they do not have to be learnt from scratch (Sanderson, 2005:45). In other words, learners do not begin the task of learning a L2 from point zero (or close to it). The authors cited above also explain that when writing or speaking the TL, L2 learners tend to rely on their L1 structures to produce a response.
However, there are currently two schools of thought pertaining to age in L2 acquisition. Some researchers (for example, Collier, 1995; and Mabila 2001) argue that learning a L2 at an early stage is advantageous as learners are confident and also have a natural ability to learn other languages, which tapers off as a person ages. Others (for example, Bailey & Buttler, 2003; Resnick, 2004
50
as well as August & Shanahan, 2006) argue that at an early (i.e. aged 4-7 years);
children take much longer to master skills needed for academic purposes and here older children are more equipped to cope. From ages 6-12, children are still in the process of developing in L1 the complex skills of reading and writing, in addition to continuing acquisition of more complex rules of morphology and syntax, elaboration of speech acts, expansion vocabulary, semantic development and even some subtleties in phonological development.
4.3.1.3 Home language of the respondents
All in all the findings from the biographical section of the data collection instrument revealed that the total percentage of learners, whose L1 was Xitsonga was 87.3%, while 1.8% revealed that TshiVenda was their home language.
Learners, who did not specify their L1, were 10.9%. The above table revealed that, the majority of the participants’ home language is Xitsonga.
With regard to L1, Sanderson (2008) notes that, a person’s L1 is usually spoken with L1 competence, while a L2 learner may never attain the same degree of proficiency as a MT speaker. Secondly, a L2 learner already has a L1 vocabulary and grammar, and does not have to construct these from scratch in the same way as a L1 learner has to. The learner’s knowledge of his or her L1 is likely to influence the way he or she approaches and learns a L2. Where the relevant features of both languages are the same, it results in correct language production known as positive transfer. The greater the differences between the two languages, the more negative the effects of the influence are likely to be.
Generally, the process will be more positive if the two languages are closer, and the more the learner is aware of the relationship between the two languages (Brown, 2000).
Table 4.1 also indicated that the learner participants revealed they also spoke other languages, such as Sepedi and English. This finding confirmed that the community, from which the sample for this study was drawn, had some level of
51
homogeneity with the rest of the country South Africa, in that it represents a multilingual society.
The following section presents data on the sources of language input, which was obtained through the questionnaire.